ICIP
2003
IEEE
14 years 6 months ago
2003
IEEE
This paper presents a new method named text to visual synthesis with appearance models (TEVISAM) for generating videorealistic talking heads. In a first step, the system learns a ...